CoATA: Effective Co-Augmentation of Topology and Attribute for Graph Neural Networks

TL;DR

CoATA is a novel dual-channel framework that enhances graph neural network performance by jointly augmenting topology and attributes through propagation, reconstruction, and contrastive learning, effectively handling noise and incompleteness in real-world graphs.

Contribution

This paper introduces CoATA, the first framework to simultaneously augment topology and attributes in GNNs, leveraging mutual reinforcement and contrastive learning for improved robustness.

Findings

Outperforms 11 state-of-the-art methods on 7 benchmark datasets.

Effectively denoises and reconstructs graph structures and attributes.

Demonstrates significant improvements in GNN robustness and accuracy.

Abstract

Graph Neural Networks (GNNs) have garnered substantial attention due to their remarkable capability in learning graph representations. However, real-world graphs often exhibit substantial noise and incompleteness, which severely degrades the performance of GNNs. Existing methods typically address this issue through single-dimensional augmentation, focusing either on refining topology structures or perturbing node attributes, thereby overlooking the deeper interplays between the two. To bridge this gap, this paper presents CoATA, a dual-channel GNN framework specifically designed for the Co-Augmentation of Topology and Attribute. Specifically, CoATA first propagates structural signals to enrich and denoise node attributes. Then, it projects the enhanced attribute space into a node-attribute bipartite graph for further refinement or reconstruction of the underlying structure. Subsequently, CoATA introduces contrastive learning, leveraging prototype alignment and consistency constraints, to facilitate mutual corrections between the augmented and original graphs. Finally, extensive experiments on seven benchmark datasets demonstrate that the proposed CoATA outperforms eleven state-of-the-art baseline methods, showcasing its effectiveness in capturing the synergistic relationship between topology and attributes.